Introduction
The 2004 House elections may have been the least competitive of the postwar era. There was almost no change in the partisan composition of the House of Representatives. Republicans gained three seats, but this was entirely the result of a redistricting plan enacted by the Republican-controlled state legislature in Texas. Outside of Texas, Republicans lost two seats in the House. Of 401 incumbents running for reelection, only five were defeated, and two of these losses were a direct result of redistricting in Texas. The 99 percent reelection rate for incumbents in 2004 equaled the success rate of incumbents in the 2002 midterm election. However a broader gauge of competition, the number of relatively close contests, shows that the 2004 House elections were even less competitive than the 2002 House elections.
Figure 1 displays the distribution of the two-party division of the vote across all 435 House districts in 2004. Only 22 contests in the entire country were decided by a margin of less than 10 percentage points. In 2002, 36 House contests were decided by a margin of less than 10 percentage points. At the other end of the competitive spectrum, 172 winning candidates in 2004 either had no major party opposition or coasted to victory by a margin of at least 40 percentage points.
Figure 1. Competitiveness of House Races in 2004
Source: Data compiled by
authors.
The 2004 House elections were extraordinarily uncompetitive. However, competition in House elections has been declining for more than 50 years. One common measure of competition in House elections is the reelection rate of incumbents. Figure 2 displays the trend in the reelection rate of House incumbents since the end of World War II. We have separated the results by region because elections in the southern states were still dominated by the Democratic Party during the first part of this time period. In the 1946, 1948, and 1950 elections, for example, not a single incumbent was defeated in the 11 states of the old Confederacy.
The results displayed in Figure 2 show that the reelection rate for incumbents in the South fell slightly between the 1940s and the 1980s with the emergence of two-party competition in the region, then rose slightly during the 1990s and the first two elections of the 21st century. During the 1980s, the reelection rate for incumbents in the South was actually slightly lower than the reelection rate for incumbents in the rest of the country. Outside of the South, the reelection rate for House incumbents increased steadily between the 1940s and the 1980s, fell slightly during the 1990s, and then rose again after 2000. For the nation as a whole, the first two elections of the 21st century produced the highest incumbent reelection rates ever recorded.
Figure
2. Proportion of House Incumbents
Reelected by Region and Decade
Source:
Data compiled by authors.
Figure 3. Proportion of Competitive House Races by
Region and Decade
Source:
Data compiled by authors.
The reelection rate for incumbents provides only a partial perspective on competition in House elections, however. Some incumbents win easily while others barely survive and not all races involve incumbents. A broader measure of competition is the proportion of relatively close contests--those decided by less than 10 percentage points or by less than 20 percentage points. Figure 3 displays the trend in the proportion of House races decided by a margin of less than 20 percentage points since the end of World War II.[1] Once again, because of the absence of two-party competition in the South during the first part of this time period, we have separated the results by region.
The results displayed in Figure 3 show that there was a gradual increase in competition in the South between the 1940s and the 1980s along with a drastic decline in competition in the rest of the country. After 1980, the proportion of competitive races was almost equal in the two regions. There was a modest increase in competition in both regions during the 1990s as Republicans regained control of the House for the first time in 50 years. After 2000, however, the proportion of competitive contests fell in both the South and the rest of the nation. Despite the relatively narrow margin by which Republicans have controlled the House of Representatives, the close division of the electorate between Democratic and Republican identifiers, and the intense competition that has characterized the last two presidential elections, the 2002 and 2004 House elections were the least competitive of the postwar era.
Explaining Declining Competition
Meaningful
competition has become increasingly rare in House elections. In the remainder of this paper we address
the question of what explains this decline in competition. We test three hypotheses that attempt to
explain declining competition in House elections: the redistricting hypothesis,
the partisan polarization hypothesis, and the incumbency hypothesis. The redistricting hypothesis, which appears
to enjoy the status of conventional wisdom among media commentators and
editorial writers, argues that declining competition is due mainly to the effects
of partisan or bipartisan gerrymandering.
According to this hypothesis, state legislatures using sophisticated new
computer-based technology have been skillfully drawing congressional district
lines to either maximize partisan gains, creating as many districts as possible
that favor the majority party while packing minority party voters into as few
districts as possible, or to protect incumbents of both parties. In either case, the results of this process,
according to the redistricting hypothesis, have been an increase in the number
of districts that are safe for one party and a decrease in the number of
marginal districts. As the New
York Times editorial page recently
opined, “both parties have succeeded in drawing district lines in ways that
cement their current power by eliminating contested elections (New York Times
2004).” Similarly, David Broder, the
dean of Washington political columnists, recently argued that district lines in
“most states” were drawn to protect incumbents from “the inconvenience of
competition (Broder 2004).”
While
the refrain in popular media outlets is nearly unanimous on redistricting, the
scholarly literature is more split.
Following Mayhew’s original observations on the “vanishing marginals,”
one political scientist found that redistricting was the cause of most of the
decline in marginal districts during the 1960s (Tufte 1973). Similarly, a statistical analysis showed
that incumbents benefited from reduced competition in 1992 as a result of
redistricting, but that this effect was evident mainly in states where the
redistricting was bipartisan (Lyons and Galderisi 1995).
On
the other hand, some scholars have argued that redistricting has a neutral
effect on competition, and may even increase it. This is primarily because in redistricting, partisans drawing
district lines face a fundamental tension between incumbent protection and
maximizing their party’s electoral potential.
More often than not, the only way to shift marginal districts toward the
party is to cut the safety margins of incumbents by moving reliable partisans
out of their districts. For this
reason, it is often the case that partisan redistricting has the effect of
reducing the safety of incumbents, thereby making elections more competitive
(Gelman and King 1994).
From a national perspective,
a study of the 1970 round of redistricting, the first in the “modern” era of
redistricting following the Supreme Court’s “one-person, one vote” rulings,
found no systematic effects on either incumbent safety or majority party
representation. The only exception was
that a few incumbents were harmed when they were placed in districts with other
incumbents (Glazer et al. 1987). Moving
to the next redistricting decade, a study of the effects of redistricting on
the 1982 elections found that party leaders appear to have traded incumbent
security for potential seat gains in drawing district lines (Gopoian and West
1984). Overall, while there is not a
consensus in the scholarly community about the consequences of redistricting
for congressional competition, the bulk of the research in this area has not
found that redistricting results in reduced competition.
Hypothesis 1: If the
redistricting hypothesis is correct, then we should observe a substantial
increase in the number of safe districts and a substantial decrease in the
number of marginal districts in the three elections that immediately followed
each recent redistricting cycle: 1982, 1992, and 2002.
The partisan polarization hypothesis offers a somewhat different explanation for the decline in competition in House elections. According to this hypothesis, House districts have become less competitive over time, but this trend is not a result of partisan gerrymandering. Instead, Democratic districts have become more Democratic, Republican districts have become more Republican, and marginal districts have been disappearing as a result of powerful social forces at work in American society, including internal migration, immigration, and ideological realignment within the electorate.
Like the redistricting theory,
the partisan polarization theory has received considerable support in the
popular press. In a cover story in The Atlantic, Brooks (2001) argued that profound cultural differences based on
education, income, religion, and numerous other factors were leading to a
growing partisan divide in the United States.
Other commentators have accepted this wisdom, sometimes coloring it with
references that further exaggerate the degree of division. One widely read columnist for The Washington Post put it this way: “the red states get redder, the
blue states get bluer, and the political map of the United States takes
on the coloration of the Civil War.” (Dionne 2003).
Some in academia have
repeated this characterization of a “deep cultural divide between the red
states and the blue states” (Mannan 2004).
One widely discussed academic work goes so far as to argue that cultural
divisions are so deep they have put America on a path toward political violence
(Hunter 1995). However, other observers
have argued that the idea of an increasingly polarized America is a “myth,” and
that in reality Americans are remarkably moderate and tolerant in their
political views (Fiorina et al. 2004).
Staking out something of a middle ground in a detailed study of voting
trends in twelve states, two political scientists make a strong case that
national political forces are moderated by contextual factors at the state
level (Gimpel and Shucknecht 2003).
According to this argument, a range of complex geographically linked
factors such as immigration, migration, education, income, and religion may be
contributing to growing geographic divergence in party loyalties.
There is also strong evidence
that growing ideological polarization at the elite level has made it easier for
voters to choose their party identification on the basis of their ideological
preferences (Abramowitz and Saunders 1998; Jacobson 2000). Therefore what many observers describe as
“polarization,” might more accurately be described as “sorting,” as voters
bring their policy and partisan preferences into alignment (Levendusky
2004). It seems likely that this
process of partisan realignment is also affecting competition in congressional
elections. Conservative districts, some
of which once regularly elected conservative Democrats, have been moving
increasingly toward the Republican Party (Black and Black 2004). At the same time, liberal districts, some of
which once regularly elected liberal Republicans, have been moving increasingly
toward the Democrats.
Hypothesis 2: If
the partisan polarization hypothesis is correct, then we should find that the
number of safe districts has been steadily increasing and the number of
marginal districts has been steadily decreasing, with most of this change
occurring between redistricting cycles.
In contrast to both the redistricting hypothesis and the partisan polarization hypothesis, the incumbency hypothesis argues that declining competition in House elections is due less to change in the partisan composition of House districts than to the growing advantage of incumbency. There are, of course, two potential sets of factors to which incumbent politicians might owe their extraordinarily high reelection rates. The first are advantages that affect their own visibility and popularity such as ex-officio resources and campaign funds. The second set of factors involves challenger resources such as political experience and campaign funds. Although the debate is ongoing, recent research points most strongly toward the second set of factors, those tied to challenger resources, as the cause of the recent spike in incumbent electoral performance.
A statistical analysis of panel data designed to decompose the sources of the incumbency advantage in House races from 1984 to 1990 puts the blame squarely on the second set of factors, and follows the “strategic politicians” hypothesis first proposed by Jacobson and Kernell (1983). Specifically, it finds that direct benefits to incumbents (e.g. ex-officio resources) account for less than half of the incumbency advantage, while the ability of incumbents to deter high quality challengers is the primary explanatory variable (Levitt and Wolfram 1997). In other words, incumbents are running their margins way up because they face low quality, underfinanced challengers. Similar results were obtained from an OLS analysis of election results during the same time period, with deterrence of quality challengers once again proving to be the primary explanation for the growth in incumbency advantage (Cox and Katz 1996). These results have found support from at least one detailed qualitative case study of potential challengers in two congressional districts (Kazee 1983). This study found that potentially strong challengers were reluctant to take on entrenched incumbents because of the long odds of success and the enormous amount of money needed to wage a competitive campaign. Another study of declining competition in House elections during the 1980s pointed to money as a critical variable: this study found that challengers were raising less money and getting less return on their campaign expenditures than they had in the previous decade (Abramowitz 1991).
Hypothesis 3: If
the incumbency hypothesis is correct, then we should find that even in marginal
districts, competition is now relatively rare because of the inability of
challengers to compete financially. We
should also find that the decline in competition in House elections has been
confined mainly to races involving incumbents, with open seat races remaining
about as competitive as in the past.
Data and Measures
We have collected data on competition in U.S. House elections since the end of World War II.[2] These data include the percentage of the major party vote won by the Democratic and Republican candidates in every House election between 1946 and 2004, the party affiliation of the incumbent Representative in each district, whether the incumbent was running for reelection, the previous political experience of the challenger, the percentage of the major party vote won by the Democratic and Republican presidential candidates in every House district between 1956 and 2004 except for the 1962 midterm election, and total campaign spending by the Democratic and Republican candidates in every House contest between 1972 and 2002.
We use the normalized presidential vote in each House district in the current or most recent presidential election as a measure of the partisan composition of the district. This
involves subtracting the Democratic or Republican presidential candidate’s percentage of the major party vote in the entire nation from his percentage of the vote in the district. For the 2004 election, we used the results of the 2000 presidential election to construct this measure because
2004 presidential election results were not yet available for individual House districts.
Findings
Since the 1994 election, Republicans have maintained a narrow majority in the House of Representatives. The number of Republican seats has ranged between 223 and 232 while the number of Democratic seats has ranged between 203 and 212. Because of the small size of the Republican majority, control of the House appears to be at stake in every election. Despite the appearance of national competitiveness, however, the number of competitive House contests has fallen since 1994, reaching record low levels in 2002 and 2004. One possible explanation for this decline in competition in individual House races is that there are fewer marginal House districts than in the past.
Figure 4. Proportion of Marginal and Safe House
Districts by Decade
Note:
Marginal districts are those in which Democratic presidential candidate’s
percentage of major party vote is within 5 points of national percentage; safe
districts are those in which Democratic presidential candidate’s percentage of
major party vote is more than 10 points above or below national percentage.
Source:
Data compiled by authors.
Figure 4 shows that during the 1950s and 1960s there were slightly more marginal districts (those in which the Democratic presidential vote was within 5 percentage points of the national average) than safe districts (those in which the Democratic presidential vote was more than 10 percentage points above or below the national average) and that during the 1970s and 1980s, the number of marginal districts was substantially greater than the number of safe districts. Between the 1980s and the 1990s, however, a dramatic change occurred: for the first time, the number of safe districts exceeded the number of marginal districts. This trend continued and accelerated during the first two elections of the 21st century. In 2002 and 2004, almost half of all House districts were classified as safe for one party or the other compared with less than one-third during the 1970s and 1980s. Meanwhile, the number of marginal districts fell from about 40 percent of the total during the 1970s to about 25 percent of the total in 2002 and 2004.
What has caused the number of marginal districts to shrink and the number of safe districts to grow since the 1970s? A number of pundits and editorial writers have pointed to partisan and bipartisan gerrymandering as an explanation for the decline in the competitiveness of House districts in recent years. According to this explanation, the decrease in the number of marginal districts and the increase in the number of safe districts are the results of carefully crafted redistricting plans designed to maximize majority party seats by packing minority party voters into a few districts as possible or to protect incumbents of both parties.
If redistricting was in fact responsible for the decline in the competitiveness of House districts, we would expect to see a substantial decrease in the number of marginal districts and a substantial increase in the number of safe districts immediately after each recent redistricting cycle. By using the presidential vote as an indicator of the partisan composition of House districts before and after redistricting, we can evaluate the impact of redistricting on the numbers of marginal and safe districts.
The evidence displayed in Figure 5 shows that redistricting did not cause a substantial increase in the number of safe districts or a substantial decrease in the number of marginal districts in 1982, 1992 or 2002. In 1982 and 2002 there were small increases in the number of safe seats. In 1992 the change was slightly larger, but it may have resulted from the creation of majority-minority districts rather than any effort to protect incumbents. Moreover, the number of marginal districts also increased between 1990 and 1992.
Figure 5. Proportion of Marginal and Safe Districts Before and
After Redistricting, 1980-2002
Note:
Marginal districts are those in which Democratic presidential candidate’s
percentage of major party vote is within 5 points of national percentage; safe
districts are those in which Democratic presidential candidate’s percentage of
major party vote is more than 10 points above or below national percentage.
Source:
Data compiled by authors.
In general, redistricting had a minor impact on the competitiveness of House districts in each of the last three redistricting cycles. Instead, the most significant changes in the competitiveness of House districts occurred between redistricting cycles. This evidence is consistent with the partisan polarization hypothesis. As a result of population movement, immigration, and ideological realignment within the electorate, Republicans are increasingly surrounded by other Republicans and Democrats by other Democrats. This trend has been evident since the 1970s, but it appears to have accelerated in recent years. Between 1992 and 2004, the number of marginal districts fell from 157 to 112 while the number of safe districts rose from 156 to 208.
The effect of this increase in partisan polarization has been magnified by the growing consistency of voting behavior between presidential and House elections. Figure 8 shows that the correlation between the Democratic percentage of the House vote and the Democratic percentage of the presidential vote in House districts has increased dramatically since the 1970s and reached an all time high in 2002-04.[3] This was not simply a result of partisan realignment in the South: the same trend is evident in congressional districts outside of the South as well as those in the South. This growing consistency appears to reflect an increase in partisan voting in both presidential and congressional elections since the 1980s (Bartels 2000).
Figure 6. Average Shared Variance Between House
and
Presidential Election Results by Decade and Region
Source:
Data compiled by authors.
As a result of both increased partisan polarization and increased partisan consistency in voting behavior, far fewer members of the House now occupy high-risk districts, districts that are less supportive of their party than the national average, and far more members now occupy low-risk districts, districts that that are at least 10 percentage points more supportive of their party than the national average. Table 1 displays the average percentage of Democratic and Republican members occupying high-risk and low-risk districts in each decade since the 1970s. During this time period, the percentage of Republicans in high-risk districts has fallen from 21.5 percent to 11.3 percent and the percentage of Democrats in high-risk districts has fallen from 41.4 percent to 16.4 percent. At the same time, the percentage of Republicans in low-risk districts has risen from 23.2 percent to 40.8 percent and the percentage of Democrats in low-risk districts has risen from 24.5 percent to 51.1 percent. Between 1972 and 2004, the total number of members occupying high risk districts fell from 157 to 97 while the total number of members occupying low risk districts rose from 50 to 203.
Table 1. Percentage of Democratic and Republican Representatives
in High-Risk and Low-Risk Districts by Decade
____________________________________________________________
1972-1980 1982-1990 1992-2000 2002-2004
____________________________________________________________
Republicans
Low-Risk 23.2 31.5 35.0 40.8
High-Risk 21.5 16.1 16.4 11.3
Democrats
Low-Risk 24.5 24.5 39.5 51.1
High-Risk 41.4 35.4 25.5 16.4
____________________________________________________________
Note: Low-risk
districts are those in which share of major party vote for presidential
candidate of incumbent’s party was at least 10 percentage points greater than
national vote share. High-risk
districts are those in which share of major party vote for presidential
candidate of incumbent’s party was less than national vote share.
Source: Data compiled
by authors.
The changing partisan composition of House districts over time has important implications for competition in House elections. Compared with 30 years ago, a much smaller proportion of members now represent districts which, based on presidential voting patterns, favor the opposing party: fewer Republicans represent Democratic-leaning districts and far fewer Democrats represent Republican-leaning districts. These high-risk districts account for a disproportionate share of incumbent defeats and party turnover in House elections. At the same time, a much larger proportion of members now represent districts which, based on presidential voting patterns, strongly favor their own party: more Republicans represent solidly Republican districts and more Democrats represent solidly Democratic districts. These low-risk districts account for a disproportionate share of uncontested and one-sided races. However, partisan polarization alone does not explain the decline in competition in House elections since the 1970s.
The partisan composition of a House district is not the only factor determining whether that district is going to have a competitive race. The data presented in Table 2 shows that even in marginal districts, those with the most even balance of party support, only a small minority of House contests in 2002 and 2004 were actually competitive. Only 11 percent of contests in these marginal districts were highly competitive, decided by a margin of less than 10 percentage points, and only 20 percent were moderately competitive, decided by a margin of between 10 and 20 percentage points. More than two-thirds of House races in marginal districts were blowouts in which the winning candidate was unopposed or received at least 60 percent of the major party vote.
Table 2. Competition in 2002-2004 House Elections by
Presidential Margin in District
_______________________________________________________
Presidential
Margin in District
____________________________________
Winner’s % of
House Vote GT 20% 10-20% LT 10%
_______________________________________________________
Uncontested
or 70+ 64% 30% 18%
60-70 28 50 52
55-60 5 10 20
50-55 3 10 11
Total 100% 100% 100%
(n) (411) (231) (228)
_______________________________________________________
Source:
Data compiled by authors.
Why were there so few competitive House races in 2002 and 2004, even in these marginal districts? One possible explanation for the lack of competition in many marginal districts is the advantage of incumbency. Because of their ex-officio advantages (Mayhew 1974), their ability to raise huge campaign war-chests, and the inability of most of their challengers to raise the funds required to mount serious campaigns, the large majority of House incumbents, even in marginal districts, win reelection by wide margins. When the advantage of incumbency is removed, however, there is a much greater chance of a competitive contest. The data displayed in Table 3 show that in these marginal districts there was a huge difference in competitiveness between contests with running incumbents and those for open seats. In districts with running incumbents, only 5 percent of the races were highly competitive and only 19 percent were moderately competitive. In contrast, in districts without incumbents, 45 percent of the races were highly competitive and another 23 percent were moderately competitive.
Table 3. Competition in Marginal
House Districts by Type of Contest, 2002-2004
_______________________________________________________
Type
of Contest
____________________________________
Winner’s % of Running
House Vote Open
Seat Incumbent
_______________________________________________________
Uncontested
or
70+ 3% 20%
60-70 29 55
55-60 23 19
50-55 45 5
Total 100% 100%
(n) (31) (197)
_______________________________________________________
Note:
Marginal districts are those in which Democratic
presidential
candidate’s percentage of major party vote is
within 5
points of national percentage.
Source: Data compiled by authors.
Figure 7 displays the trend in the percentage of competitive open seat and incumbent races since the end of World War II. In the first three elections of the postwar era there was very little difference in the competitiveness of open seat and incumbent races. Since the 1950s, however, open seat House races have generally been much more competitive than those with incumbents. Moreover, while the extent of competition for open seats has changed very little, there has been a fairly steady decline in competition for seats in which incumbents were running. During the 1950s, about 40 percent of races with incumbents were at least moderately competitive. In the first two elections of the 21st century, however, only about 15 percent of races with incumbents were at least moderately competitive. The advantage of incumbency seems to be growing, but why?
Figure 7. Competition in Incumbent and Open Seat
House Races by Decade
Source:
Data compiled by authors.
While the ex-officio advantages of incumbency have changed very little since the 1970s, the ability of incumbents to dominate their challengers financially has continued to increase. Figures 8 and 9 both show the trend in the percentage of challengers who were financially competitive—accounting for at least 40 percent of total spending—in House elections between 1972 and 2002. In Figure 8 the trend is broken down by type of district. In Figure 9 the trend is broken down by the political experience of the challenger.
Figure 8. Financial Competitiveness of House
Challengers by
Type of District, 1972-2002
Note: Challengers classified as financially competitive if
they accounted for at least 40 percent of combined challenger and incumbent
spending. High-risk districts are those
in which vote percentage of vote for presidential candidate of incumbent’s
party is less than national percentage.
Source:
Data compiled by authors.
Figure
9. Financial Competitiveness of House Challengers by Political Experience,
1972-2002
Note: Experienced challengers are those who have held any
elected office.
Source:
Data compiled by authors.
The results displayed in Figures 8 and 9 show that even among challengers in potentially competitive districts and even among challengers with previous office-holding experience, there has been a fairly drastic decline in financial competitiveness since the 1970s. The results displayed in Figure 8 show that there has been a sharp decline in the percentage of financially competitive challengers in high risk districts—those in which the party balance was most favorable to the challenger. The results displayed in Figure 9 show that the decline in the percentage of financially competitive challengers has been even greater among politically experienced challengers than among politically inexperienced challengers. Taken together, these results suggest that even those challengers with the greatest potential for running competitive campaigns are having more and more difficulty raising the funds necessary to wage such campaigns.
Table 4. Competition in High-Risk Incumbent Districts by Challenger’s Share of Spending, 1992-2002
_______________________________________________________
Challenger’s
% of Spending
_________________________________
Challenger’s
% of Vote LT 20 20-40 GT 40
_______________________________________________________
LT
30 22% 3% 0%
30-40 55 23 12
40-45 15 32 26
45
+ 7 42 62
Total 100% 100% 100%
(n) (175) (172) (103)
_______________________________________________________
Note:
High-risk districts are those in which percentage of major party vote for
presidential candidate of incumbent’s party is less than national percentage.
Source: Data compiled by authors.
The ability to dominate their challengers financially can shield incumbents who represent high-risk districts from defeat or even from close calls. The results displayed in Table 4 show that in elections between 1992 and 2002 in high-risk districts—those with the most unfavorable partisan composition for the incumbent—there was a very strong relationship between the challenger’s share of campaign spending and the competitiveness of the race. Sixty-two percent of contests in which the challenger accounted for at least 40 percent of total spending were highly competitive. In contrast, only 7 percent of contests in which the challenger accounted for less than 20 percent of total spending were highly competitive.
One of the main reasons for the decline in financial competitiveness among challengers is that the cost of waging a competitive campaign against an incumbent has increased dramatically in the past decade. Figure 10 displays the trend in the average expenditures of challengers in moderately and highly competitive House races for each presidential-midterm election cycle since the 1970s. Between the 1994-1996 election cycle and the 2000-2002 election cycle, the average amount spent by challengers in moderately competitive races rose from $246,000 to $699,000; during this same period, the average amount spent by challengers in highly competitive races rose from $454,000 to $1,289,000.
Figure 10. Average Expenditures of Challengers in Moderately
and Highly Competitive Races by Election Cycle
As the cost of waging a competitive campaign has soared, the proportion of challengers capable of raising this kind of money has fallen. During the 1992-1994 election cycle, 26 percent of challengers spent at least $246,000 and 12 percent spent at least $454,000; during the 2000-2002 election cycle, only 13 percent of challengers spent at least $699,000 and only 6 percent spent at least $1,289,000. Even in potentially competitive districts, those in which the presidential candidate from the incumbent’s party received less than his national share of the vote, the proportion of challengers able to raise the amount needed to wage a competitive campaign has fallen dramatically in the past decade. During the 1992-1994 election cycle, 46 percent of challengers in these potentially competitive districts raised at least $246,000 and 24 percent raised at least $454,000; during the 2000-2002 election cycle, only 30 percent of challengers in potentially competitive districts raised at least $699,000 and only 12 percent raised at least $1,289,000. These results indicate that the rising cost of congressional campaigns is becoming a larger and larger obstacle to competition in House elections.
Discussion and Conclusions
The evidence presented in this paper indicates that declining competition in U.S. House elections is explained by two major factors: a shift in the partisan composition of House districts and a decline in the ability of challengers to compete financially with incumbents. Since the 1970s, and especially since 1992, there has been a substantial increase in partisan polarization among House districts. The number of marginal districts has been declining while the number of districts that are safe for one party has been increasing. Redistricting appears to have little or nothing to do with this trend: almost all of the change in district partisanship has occurred between redistricting cycles.
Along with the increasing polarization of House districts, there has been a substantial increase in partisan voting: voting in House elections is now much more consistent with voting in presidential elections. As a result of both of these trends, districts held by Democrats are now more strongly Democratic than in the past and districts held by Republicans are now more strongly Republican than in the past. There are far fewer Democrats and Republicans occupying high-risk districts, districts whose partisan composition favors the opposing party, and far more Democrats and Republicans occupying low-risk districts, districts whose partisan composition strongly favors their own party.
The increasing polarization of Democratic and Republican districts has important implications for governance and for congressional elections. The decreasing proportion of Democrats and Republicans representing marginal or high-risk districts means that there are fewer members who have an incentive to cross party lines on issues in order to appeal to supporters of the opposing party in their districts. The result is increased party-line voting and reduced bipartisan cooperation on major issues. Increasing polarization also means that both parties have fewer seats at risk in elections. The result is diminished seat turnover and smaller seat swings. This helps to explain why, despite the relatively small size of their majority, Republicans have been able to maintain control of the House in five consecutive elections since their 1994 takeover.
The effects of increasing partisan polarization have been
reinforced by the second trend uncovered by our study—the decreasing financial
competitiveness of House challengers.
Not only are there fewer incumbents in high-risk districts, but even in
these districts, incumbents running for reelection are less likely to face
financially competitive challengers.
Fewer and fewer challengers are able to raise the amount of money that
is now required to wage a competitive campaign against a well-funded incumbent. As a result, competition is now confined to
open seats and a handful of races involving exceptionally vulnerable incumbents
and/or exceptionally well-financed challengers.
References
Abramowitz, Alan I. 1991. “Incumbency, Campaign Spending, and the Decline of Competition in U.S. House Elections.” Journal of Politics. 53: 34-56.
Abramowitz, Alan I.
and Kyle Saunders. 1998. “Ideological Realignment in the U.S. Electorate.” Journal of Politics. 60: 634-652.
Bartels, Larry
M. 2000. “Partisanship and Voting Behavior, 1952-1996.” American Journal of Political Science. 44: 35-50.
Black, Earl, and
Merle Black. 2004. The Rise of Southern Republicans. Cambridge,
MA: Harvard University Press.
Broder, David. 2004.
“No Vote Necessary.” The Washington Post. November 11: A37.
Brooks, David. 2001.
“One Nation, Slightly Divisible.” The
Atlantic Monthly. December: 53-65.
Cox, Gary W. and Jonathan N. Katz. 1996. “Why Did the Incumbency Advantage in U.S. House Elections Grow?” American Journal of Political Science. 40: 478-497.
Dionne, E.J. 2003.
“One Nation Deeply Divided.” The
Washington Post. November 7: A3.
Fiorina, Morris P.,
Samuel Abrams and Jeremy C. Pope.
2004. Culture War? The
Myth of a Polarized America. New York: Person Longman.
Gelman, Andrew and
Gary King. 1994. “Enhancing Democracy through Legislative Redistricting.” American
Political Science Review. 88: 541-559.
Gimpel, James G. and
Jason E. Schuknecht. 2003. Patchwork Nation:
Sectionalism and Political Change in American Politics. Ann Arbor: University of Michigan Press.
Glazer, Amihai,
Bernard Grofman, and Marc Robbins. 1987. “Partisan and Incumbency Effects of
1970 Redistricting.” American Journal of Political Science. 3: 680-707.
Gopoian, J. David and Darrell M. West. 1984. “Trading Security for Seats:
Strategic Considerations in the Redistricting Process.” Journal of Politics.
48: 1080-1096.
Hunter, James
Davison. 1995. Before the
Shooting Begins: Searching for Democracy in America’s Culture Wars. New York: Free Press.
Jacobson, Gary C.
2000. “Party Polarization in National Politics: The Electoral Connection.”
In Polarized Politics:
Congress and the President in a Partisan Era. Jon R. Bond and Richard Fleisher, eds. Washington, DC: Congressional Quarterly Press.
Jacobson, Gary C. and Samuel Kernell. 1983. Strategy and Choice in Congressional Elections. New Haven: Yale.
Kazee, Thomas A. 1983. “The Deterrent Effect of Incumbency on Recruiting Challengers in U.S. House Elections.” Legislative Studies Quarterly. 8: 469-480.
Levendusky, Matthew
S. 2004. “Sorting, Not Polarization: The Changing Nature of Party ID and
Ideology in the U.S. Electorate.” Paper presented at the Annual Meeting of the
American Political Science Association, Chicago, IL.
Levitt, Steven D. and Catherine D. Wolfram. 1997. “Decomposing the Sources of Incumbency Advantage in the U.S. House.” Legislative Studies Quarterly. 22: 45-60.
Lyons, Michael and
Peter F. Galderisi. 1995. “Incumbency, Reapportionment and U.S. House
Redistricting.” Political Research Quarterly. 48: 857-871.
Mannan, Rossana.
2004. “UCI Professors Examine Political Divide Among States.” New
University.
November 8: News.
Mayhew, David R. 1974.
Congress: The Electoral
Connection. New Haven: Yale University Press.
New York Times. 2004.
“Elections with No Meaning.” The New York Times. February 21.
Tufte, Edward R.
1973. “The Relationship Between Seats and Votes in Two Party Systems.” American
Political Science Review. 67: 540-555.
Weber, Ronald E.,
Harvey J. Tucker, and Paul Brace. 1991. “Vanishing Marginals in State
Legislative Elections.” Legislative Studies Quarterly. 16: 29-47.